^j6 SECOND A RV CHANGES. 



younger stem and apical branches of Pinus silvestris and its allies, the cork layers 

 which cut off the young thin scales of bark consist, on the outside and inside 

 respectively, of a few-layered stratum of thin-walled, easily torn cells, while between 

 them a stratum occurs which consists of one or two layers of sclerotic elements ; 

 this does not always extend to the edge of the thin-walled strata. When the latter 

 become torn the whole bark peels away, and forms partly the somewhat thin scales 

 of true bark, consisting of desiccated cortical tissue and adherent layers of cork, 

 partly those tough feathery sheets, of the thickness of paper, which are the sclerotic, 

 hard, persistent layers of the suberous zones. 



The peeling annular bark of species of INIelaleuca (especially M. styphelioides), 

 Callisteraon, Vitis, Clematis, &c., also, strictly speaking, belongs to this category. It is 

 however held fast for a long time in the form of fibrous flaps adhering to the cortex, 

 and hindered from falling off by the numerous strands of bast-fibres which belong 

 to the cortical tissue in process of desiccation; they traverse and support each 

 desquamated zone surrounding the periphery of the stem, as a strong network 

 of fibres with pointed meshes. 



In cases where the cohesion of the old suberous membranes is greater, the 

 layers of bark succeeding one another from without inwards adhere more closely 

 one to another, as a connected crust, which, as growth in thickness goes on, becomes 

 more and more cracked externally, and gradually suffers decay. Woody plants with 

 a thick, cracked, and rugged bark, such as Oaks, Birches, Poplars, most Willows, 

 Robinia, &c., afford universally known examples of this phenomenon; the cortex of the 

 old stem of Pinus sylvestris may also be expressly mentioned as an example, because 

 the difference which it so conspicuously shows from that of young stems and branches 

 depends principally, though not exclusively, on the different cohesion of the above- 

 mentioned thin-walled layers of cork. The further differences consist in the structure 

 of the desquamating layer of bast, which is no doubt altered by the extensive 

 formation of phelloderm, and in other respects still requires further study. 



A necessary direct connection between the direction of the cracks in the surface, 

 and the form, size, &c., of the scales of bark cut off, does not exist, at least 

 not in the typical cases belonging to this series. In those which are intermediate 

 between the latter and the cases of complete desquamation, represented by the Plane, 

 such a relation may occur. 



. Secondly, and independently of the cohesion of the membranes, the thickness and 

 special structure of the entire suberous layer come under consideration, and in this 

 respect essentially the same differences exist, as were stated above, with regard 

 to the formation of cork in general, and to superficial periderms. 



In most cases, no doubt, the layers of cork which cut off the bark are mem- 

 branes, a few (not much more than ten) layers of cells in thickness; the cells 

 themselves then either belong to the flat form, e. g. Platanus and Pinus sylvestris, 

 or are wide and even radially elongated (e. g. IVIelaleuca), the successive layers being 

 similar or dissimilar. 



On the other hand, in many cases of repeated internal formation of 

 periderm, cork is produced in large masses, forming thick strata, consisting of 

 very many layers, and perceptible to the naked eye as broad zones. In these cases the 

 cork no doubt always belongs to the wide-celled, thin-walled form, or consists of 



